Prevalence of thoracic degenerative MRI findings and association with pain and disability: a systematic review.

Abstract

To systematically review the literature on the prevalence of degenerative MRI findings in the thoracic spine and their association with pain and disability. The Medline, EMBASE, CINAHL, and CENTRAL databases were searched. Two independent reviewers screened the articles, extracted the data, and assessed the risk of bias (RoB) using a modified version of the Hoy tool for articles on prevalence and QUADAS-2 for articles on associations. The review included 23 articles describing 20 study populations: 5 general populations, 2 populations of asymptomatic volunteers, and 13 clinical populations. As the articles were highly heterogeneous, especially regarding study populations and the MRI findings assessed, a meta-analysis was not possible, and the results were presented descriptively. Prevalence estimates were reported in 21 articles, and 5 (3 study populations) had low RoB. The most frequently reported prevalence rates concerned the intervertebral discs, vertebral endplates, and the ligamentum flavum. Estimates varied considerably, with the prevalence of thoracic disc degeneration ranging from 0.2%-89% and vertebral endplate signal changes ranging from 0%-82%, depending on MRI definitions, population characteristics, and recruitment settings. Four articles investigated the association between MRI findings and spinal pain, with the RoB mainly considered low or unclear. Only two of these studies specifically examined thoracic spinal pain. None assessed the association with disability. Due to heterogeneity, few high-quality studies and few studies on individual MRI findings, no definitive conclusions could be drawn about the prevalence of thoracic degenerative MRI findings or their association with pain and disability.